What is Files SDK?

Files SDK is a unified storage SDK that provides a single, consistent API for interacting with object and blob storage backends. Instead of learning a different SDK for every cloud provider, you get one small, honest interface that works across S3, Cloudflare R2, Google Cloud Storage, Azure Blob, and over 30 other providers. It uses web-standards I/O — accepting File, Blob, ReadableStream, ArrayBuffer, and string — and runs anywhere fetch runs, including Node, Bun, Workers, and Vercel.

Who it's for

• Full-stack developers who need to swap storage providers without rewriting every upload and download call in their application.
• Platform engineers building multi-tenant or multi-region systems that must abstract away provider-specific quirks behind a single class.
• Indie hackers and startup teams who want to start with a simple filesystem adapter and later migrate to S3 or R2 without changing their application code.

Key features

One small API across providers

Upload, download, list, delete, head, exists, copy, and URL generation — the same ten methods work identically whether you're using S3, GCS, Azure, or a local filesystem. Swap your adapter, and your calls stay the same.

Web-standards I/O

The SDK accepts native web types like File, Blob, and ReadableStream, and returns them too. No proprietary stream wrappers or buffer conversions. It runs on any runtime that supports the fetch API — Node, Bun, Cloudflare Workers, Vercel Edge, and more.

Escape hatch via files.raw

When you need provider-specific features like versioning, lifecycle policies, ACLs, or multipart uploads, the native client is always one property away. is typed per adapter, so you get full autocompletion for the underlying SDK without leaving your codebase.